Advances and challenges in foundation agents. From brain inspired intelligence to evolutionary, collaborative and safe systems. Liu, B., Li, X., & Zhang, J. ![Advances and challenges in foundation agents. From brain inspired intelligence to evolutionary, collaborative and safe systems [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex The advent of large language models (LLMs) has catalyzed a transformative shift in artificial intelligence, paving the way for advanced intelligent agents capable of sophisticated reasoning, robust perception, and versatile action across diverse domains. As these agents increasingly drive AI research and practical applications, their design, evaluation, and continuous improvement present intricate, multifaceted challenges. This survey provides a comprehensive overview, framing intelligent agents within a modular, brain-inspired architecture that integrates principles from cognitive science, neuroscience, and computational research. We structure our exploration into four interconnected parts. First, we delve into the modular foundation of intelligent agents, systematically mapping their cognitive, perceptual, and operational modules onto analogous human brain functionalities, and elucidating core components such as memory, world modeling, reward processing, and emotion-like systems. Second, we discuss self-enhancement and adaptive evolution mechanisms, exploring how agents autonomously refine their capabilities, adapt to dynamic environments, and achieve continual learning through automated optimization paradigms, including emerging AutoML and LLM-driven optimization strategies. Third, we examine collaborative and evolutionary multi-agent systems, investigating the collective intelligence emerging from agent interactions, cooperation, and societal structures, highlighting parallels to human social dynamics. Finally, we address the critical imperative of building safe, secure, and beneficial AI systems, emphasizing intrinsic and extrinsic security threats, ethical alignment, robustness, and practical mitigation strategies necessary for trustworthy real-world deployment. By synthesizing modular AI architectures with insights from different disciplines, this survey identifies key research gaps, challenges, and opportunities, encouraging innovations that harmonize technological advancement with meaningful societal benefit. The project’s Github link is: https://github.com/FoundationAgents/awesome-foundation-agents.
@misc{liu_advances_nodate,
title = {Advances and challenges in foundation agents. {From} brain inspired intelligence to evolutionary, collaborative and safe systems},
url = {https://arxiv.org/pdf/2504.01990},
abstract = {The advent of large language models (LLMs) has catalyzed a transformative shift in artificial intelligence, paving the way for advanced intelligent agents capable of sophisticated reasoning, robust
perception, and versatile action across diverse domains. As these agents increasingly drive AI
research and practical applications, their design, evaluation, and continuous improvement present
intricate, multifaceted challenges. This survey provides a comprehensive overview, framing intelligent
agents within a modular, brain-inspired architecture that integrates principles from cognitive science,
neuroscience, and computational research. We structure our exploration into four interconnected
parts. First, we delve into the modular foundation of intelligent agents, systematically mapping
their cognitive, perceptual, and operational modules onto analogous human brain functionalities, and
elucidating core components such as memory, world modeling, reward processing, and emotion-like
systems. Second, we discuss self-enhancement and adaptive evolution mechanisms, exploring how
agents autonomously refine their capabilities, adapt to dynamic environments, and achieve continual
learning through automated optimization paradigms, including emerging AutoML and LLM-driven
optimization strategies. Third, we examine collaborative and evolutionary multi-agent systems,
investigating the collective intelligence emerging from agent interactions, cooperation, and societal
structures, highlighting parallels to human social dynamics. Finally, we address the critical imperative
of building safe, secure, and beneficial AI systems, emphasizing intrinsic and extrinsic security
threats, ethical alignment, robustness, and practical mitigation strategies necessary for trustworthy
real-world deployment. By synthesizing modular AI architectures with insights from different disciplines, this survey identifies key research gaps, challenges, and opportunities, encouraging innovations
that harmonize technological advancement with meaningful societal benefit. The project’s Github
link is: https://github.com/FoundationAgents/awesome-foundation-agents.},
language = {Inglés},
urldate = {2025-04-08},
author = {Liu, Bang and Li, Xinfeng and Zhang, Jiayi},
keywords = {agentes},
}
Downloads: 0
{"_id":"ap8mCTHNnqs7fPrew","bibbaseid":"liu-li-zhang-advancesandchallengesinfoundationagentsfrombraininspiredintelligencetoevolutionarycollaborativeandsafesystems","author_short":["Liu, B.","Li, X.","Zhang, J."],"bibdata":{"bibtype":"misc","type":"misc","title":"Advances and challenges in foundation agents. From brain inspired intelligence to evolutionary, collaborative and safe systems","url":"https://arxiv.org/pdf/2504.01990","abstract":"The advent of large language models (LLMs) has catalyzed a transformative shift in artificial intelligence, paving the way for advanced intelligent agents capable of sophisticated reasoning, robust perception, and versatile action across diverse domains. As these agents increasingly drive AI research and practical applications, their design, evaluation, and continuous improvement present intricate, multifaceted challenges. This survey provides a comprehensive overview, framing intelligent agents within a modular, brain-inspired architecture that integrates principles from cognitive science, neuroscience, and computational research. We structure our exploration into four interconnected parts. First, we delve into the modular foundation of intelligent agents, systematically mapping their cognitive, perceptual, and operational modules onto analogous human brain functionalities, and elucidating core components such as memory, world modeling, reward processing, and emotion-like systems. Second, we discuss self-enhancement and adaptive evolution mechanisms, exploring how agents autonomously refine their capabilities, adapt to dynamic environments, and achieve continual learning through automated optimization paradigms, including emerging AutoML and LLM-driven optimization strategies. Third, we examine collaborative and evolutionary multi-agent systems, investigating the collective intelligence emerging from agent interactions, cooperation, and societal structures, highlighting parallels to human social dynamics. Finally, we address the critical imperative of building safe, secure, and beneficial AI systems, emphasizing intrinsic and extrinsic security threats, ethical alignment, robustness, and practical mitigation strategies necessary for trustworthy real-world deployment. By synthesizing modular AI architectures with insights from different disciplines, this survey identifies key research gaps, challenges, and opportunities, encouraging innovations that harmonize technological advancement with meaningful societal benefit. The project’s Github link is: https://github.com/FoundationAgents/awesome-foundation-agents.","language":"Inglés","urldate":"2025-04-08","author":[{"propositions":[],"lastnames":["Liu"],"firstnames":["Bang"],"suffixes":[]},{"propositions":[],"lastnames":["Li"],"firstnames":["Xinfeng"],"suffixes":[]},{"propositions":[],"lastnames":["Zhang"],"firstnames":["Jiayi"],"suffixes":[]}],"keywords":"agentes","bibtex":"@misc{liu_advances_nodate,\n\ttitle = {Advances and challenges in foundation agents. {From} brain inspired intelligence to evolutionary, collaborative and safe systems},\n\turl = {https://arxiv.org/pdf/2504.01990},\n\tabstract = {The advent of large language models (LLMs) has catalyzed a transformative shift in artificial intelligence, paving the way for advanced intelligent agents capable of sophisticated reasoning, robust\nperception, and versatile action across diverse domains. As these agents increasingly drive AI\nresearch and practical applications, their design, evaluation, and continuous improvement present\nintricate, multifaceted challenges. This survey provides a comprehensive overview, framing intelligent\nagents within a modular, brain-inspired architecture that integrates principles from cognitive science,\nneuroscience, and computational research. We structure our exploration into four interconnected\nparts. First, we delve into the modular foundation of intelligent agents, systematically mapping\ntheir cognitive, perceptual, and operational modules onto analogous human brain functionalities, and\nelucidating core components such as memory, world modeling, reward processing, and emotion-like\nsystems. Second, we discuss self-enhancement and adaptive evolution mechanisms, exploring how\nagents autonomously refine their capabilities, adapt to dynamic environments, and achieve continual\nlearning through automated optimization paradigms, including emerging AutoML and LLM-driven\noptimization strategies. Third, we examine collaborative and evolutionary multi-agent systems,\ninvestigating the collective intelligence emerging from agent interactions, cooperation, and societal\nstructures, highlighting parallels to human social dynamics. Finally, we address the critical imperative\nof building safe, secure, and beneficial AI systems, emphasizing intrinsic and extrinsic security\nthreats, ethical alignment, robustness, and practical mitigation strategies necessary for trustworthy\nreal-world deployment. By synthesizing modular AI architectures with insights from different disciplines, this survey identifies key research gaps, challenges, and opportunities, encouraging innovations\nthat harmonize technological advancement with meaningful societal benefit. The project’s Github\nlink is: https://github.com/FoundationAgents/awesome-foundation-agents.},\n\tlanguage = {Inglés},\n\turldate = {2025-04-08},\n\tauthor = {Liu, Bang and Li, Xinfeng and Zhang, Jiayi},\n\tkeywords = {agentes},\n}\n\n\n\n","author_short":["Liu, B.","Li, X.","Zhang, J."],"key":"liu_advances_nodate","id":"liu_advances_nodate","bibbaseid":"liu-li-zhang-advancesandchallengesinfoundationagentsfrombraininspiredintelligencetoevolutionarycollaborativeandsafesystems","role":"author","urls":{"Paper":"https://arxiv.org/pdf/2504.01990"},"keyword":["agentes"],"metadata":{"authorlinks":{}},"downloads":0,"html":""},"bibtype":"misc","biburl":"https://bibbase.org/zotero-group/elpezflaco/5927890","dataSources":["HJYZSatsNGbCiWtJj"],"keywords":["agentes"],"search_terms":["advances","challenges","foundation","agents","brain","inspired","intelligence","evolutionary","collaborative","safe","systems","liu","li","zhang"],"title":"Advances and challenges in foundation agents. From brain inspired intelligence to evolutionary, collaborative and safe systems","year":null}